DE602005002703T2 - Pulse radio systems with multiple pulse types - Google Patents

Description

BACKGROUND OF THE INVENTION

1. Field of the invention

The The present invention relates generally to radio communication systems and more extreme broadband systems (UWB).

2. Description of the state of the technique

With the publication from "First Report and Order " the federal commission for The Message Industry (FCC) on February 14, 2002 has the interest on extremely broadband systems (UWB) increased significantly. UWB Systems spread information over a wide range of at least 500 MHz. Because of this dissemination effect is the spectral power density small. Therefore, the interference is too recipients also narrow with narrow bandwidth.

transmitter for extreme broadband Transmitted-Reference Systems (TR-UWB) with low Data transfer rates are described by R. Hoctor and H. Tomlinson, "Delay-hopped transmitted-reference RF communications "(" Transmitted-Reference Radio Frequency Telecommunications with delay hopping "), Proceedings of the IEEE Conference of Ultra Wideband Systems and Technologies 2002 (Session Reports of the IEEE Conference on Extreme Broadband Systems and Technologies 2002 (UWBST '02), pp. 265-269, May 2002); N. v. Stralen, A. Dentinger, K. Welles II, R. Gaus, R. Hoctor and H. Tomlinson, "Delay hopped transmitted reference experimental results "(" Test results from Transmitted-Reference with Delay Hopping "), Proceedings of the IEEE Conference of Ultra Wideband Systems and Technologies 2002 (Session Reports the IEEE conference on extremely broadband systems and technologies 2002 (UWBST '02), Pp. 93-98, May 2002; F. Tufvesson and A.F. Molisch, "Ultra-wideband communication using hybrid matched filter correlation receivers "(" Extrem broadband communication using hybrid receivers with Comb Filter Correlation "), Proc. IEEE Vehicular Technology Conference VTC 2004 Spring (Session Reports the IEEE Conference on Vehicle Technology (VTC, Spring 2004), Milan, Italy, 17.-19. May 2004, and J.D. Choi and W.E. Strong, "Performance of ultra-wideband communications with suboptimal receivers in multipath channels "(" Performance extremely broadband communication with suboptimal receivers in Multipath Channels ", IEEE Journal Selected Areas in Communications (IEEE Journal for Selected Areas of Telecommunications), Vol. 20, Issue 9, pp. 1754-1766, December 2002.

These Systems with low data transfer rates relax the strict time requirements of impulse radio systems (IR), M.Z. Win, R.A. Scholtz, "Impulse radio: How it works ", IEEE Communications Letters, 2 (2): Pp. 36-38, Feb. 1998, and require no channel rating. Channel rating is a challenging task for coherent UWB Receiver; Lottici, A. D'Andrea and U. Mengali, "Channel estimation for ultra-wideband communications "(" Channel rating for extreme broadband communication "), IEEE Journal to Selected Areas in Communications (IEEE Journal for Selected Areas Telecommunications), Vol. 20, Issue 9, pp. 1638-1645, December 2002.

in the The prior art discloses two basic receiver schemes, namely rake receivers matched filters, see Choi et al .; and a transmitted-reference scheme, using a pulse correlator, see Hoctor et al., Delayhopped transmitted reference RF communications "(" Transmitted Reference Radio Frequency Telecommunications with delay hopping "), IEEE Conf. at Ultra Wideband Systems and Technologies (IEEE Conference to extremely broadband systems and technologies), pp. 265-270, 2002.

The Rake solution makes a channel rating for combining a selected one Number of reusable components required. Because the receiver structure fairly complex, only the strongest or some of the strongest reusable components become used to form the decision variable. It means that the recipient not all reusable components are fully analyzed and the performance due to the adherent channel rating and the combination problem less than ideal. An increase in the Number of rake fingers increased the complexity and costs of the system.

In transmitted-reference schemes, pairs of transmitted pulses are used for each data symbol. The first pulse, called the reference pulse, is not modulated by the data symbol. The second pulse, called the data pulse, is modulated by the data symbol. The reference and date n pulses are separated by a time delay. The receiver recovers the data symbol by multiplying timed pulses, resulting in a large correlation peak. The different peak values all have the same phase. The phase is determined by the value of the data symbol, and therefore it is an advantage that they can be summed by an integrator. This scheme is less complex and able to combine the energy of different multipath components without a channel rating. Unfortunately, the output of the multiplier has a very poor signal-to-noise ratio (SNR) due to nonlinear operations in terms of noise terms when the decision variable is formed and due to its own energy loss when the reference pulse is transmitted. This results in large Noise-Times-Noise terms that are integrated over time. The received signal may be passed through a matched filter to reduce the effects of noise-time noise terms, see Tufvesson et al., Or averaging may be performed, see Choi et al. Reference scheme in comparison to the ideal rake solution due to the noise products on a worse performance.

in the The state of the art is for all transmitted between a given transmitter and receiver Signals a single pulse type used. That is, all reference pulses and all data pulses are of the same type, e.g. are all a Gaussian or alternatively a Gaussian Mono cycle. By doing that for all signals have the same type of pulse is the probability Interframe Interference (IFI) and Multi-Access Interference (MAI) elevated. Therefore, there is a need for a UWB system that reduces IFI and MAI.

SUMMARY OF THE INVENTION

The The invention provides a method and system for transmitting and receiving Radio signals in an extremely broadband multi-pulse communication system with transmitted reference pulse ready. To transfer of a data symbol, a transmitter sends multiple reference pulses, each of a different type, and multi-digit data pulses, in terms of number and types of number and types correspond to the reference pulses. The pulses become one predetermined structure sent by the number and types the impulses is. By using several different types of pulses interference is reduced.

Instead of sending a reference pulse and a data pulse in each data block as in the prior art, the invention first transmits a set of N _p different reference pulses in the first N _p data blocks. Subsequently, a set of N _p different data pulses are sent in the next N _p data blocks, and so on, with the sets of reference and data pulses alternating.

The receiver processes the received signal by N _p parallel branches. Each branch acquires information from the pulses of a particular type by using template signals specific to each pulse type. The decision variables for the N _p parallel branches are combined to evaluate the output data symbol of the receiver.

The inventive system facilitates the use of multiple different pulses for improved noise suppression. Complexity constraints, interference effects, and the signal-to-noise ratio of the system are the major factors to consider in selecting the number of different pulse types N _p .

Because for different Pulse types are used different branches, it is now possible, different Combination schemes such as combining or the Weighting of equal transmission factors according to the types of impulses to use and the difficult decisions and majority decisions for contributions from different Adapting pulse types.

BRIEF DESCRIPTION OF THE DRAWINGS

preferred versions The present invention will be described based on the following figures in detail described in which:

1 is the block diagram of a transmitter according to the invention;

2 is the block diagram of a receiver according to the invention;

3 is the block diagram of another receiver according to the invention; and

4 is the Zeitstaffelungsdiagramm a radio signal for an inventive broadband system.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

signal structure

wobei N_f die Anzahl von Impulsen je Informationsbit oder Datensymbol, N_p die Anzahl von unterschiedlichen Impulstypen mit n einem Index für den Impulstyp und i einem Index für das Datensymbol ist. Der Einfachheit der Ausdrücke halber wird angenommen, dass die Anzahl N_f von Impulsen ein geradzahliges Mehrfaches der Anzahl von Datensymbolen ist. Dies ist jedoch nicht erforderlich. Die Variable s_n,i(t) wird ausgedrückt als

A transmitted signal s (t) for an extremely wideband system (UWB) according to the invention is expressed as:

where N _{f is} the number of pulses per information bit or data symbol, N _{p is} the number of different pulse types with n an index for the pulse type and i an index for the data symbol. For simplicity of expression, it is assumed that the number N _f of pulses is an even multiple of the number of data symbols. This is not necessary. The variable s _{n, i} (t) is expressed as

In equation (2), ω _n (t) is the UWB pulse of type n, T _f the data block interval and T _c the chip interval. A time-hopping code (TH) is denoted by c _j , which is an integer taking values of the set {0, 1, ... N _c -1}, where N _{c is} the number of chips per data block , This prevents devastating collisions between different users.

The random polarity, or spreading code d _j ∈ {-1, +1}, changes the polarities of the pulses, which smoothes the spectral power density of the transmitted signal; Y.-P. Nakache and AF Molisch, "Spectral shape of UWB signals of modulation format, multiple access scheme and pulse shape", session reports of the IEEE Vehicle Technology Conference, (VTC Spring 2003), Vol. 4, pp. 2510-2514, April 2003, and causes parameter insensitivity to multi-port interference (MAI); E. Fishler and HV Poor, "On the tradeoff between two types of processing gain", session reports from the 40th Annual Allerton Conference on Communication, Control and Computing, Monticello, Italy, 2 October of 2002.

The information bit or data symbol is denoted by b ⌊ 2N _p j / N _f ⌋ ∈ {-1, +1} and executed by the data pulses. In other words, the second pulse in equation (2) represents the data pulse while the first pulse is the reference pulse. T _n specifies the distance between the reference and data pulses for the nth pulse type.

There are a number of different pulse types known, for example, Gaussian, Gaussian monocycle, Gaussian dipole, etc., see U.S. Patent 6,614,384 , "System and Method for Detecting an Intruder by Pulse Radio Technology, issued to Hall et al., September 2, 2003.

4 shows exemplary signals 400 according to the invention. It should be noted that the pulses in 4 actually have three different shapes as indicated by ω ₀ , ω ₁ , ω ₂ , see U.S. Patent 6,614,384 for exemplary pulse types. In 4 are three different reference pulses 401 and three corresponding data pulses 402 present, ie N _p = 3.

For the signal, as shown, the number of data blocks is 403 (N _f ) twelve, ie the sequence of six pulses 401 - 402 is repeated twice. 4 shows only the first eight of twelve data blocks. The number of chips (N _c ) 404 in each data block is eight, and T _n = ΔT _c for n = 1, 2, 3 with Δ = 12. The time-hopping sequence is {5, 4, 2, 1, 2, 0}. For the sake of simplicity, no polarity codes are shown, ie d _j = 1 ∀j. In the example, the data symbol is -1 ', and the first, second, third, seventh, eighth, and ninth pulses are the reference pulses, while the remaining pulses are the data pulses.

The signals 411 - 413 are template pulses used in a receiver according to the invention, see 2 and 3 below.

transmitter structure

Vor der Verstärkung 140, 150 und einer Übertragung über die Antenne 160 wird ein Polaritätscode 130 auf die Impulsfolgen angewandt. 1 shows a transmitter according to the invention 100 , A data source 101 generates data symbols. Multiple copies 102 - 109 The data symbols are of the control logic 110 and the pulse generator 120 dependent. There is one copy for each pulse type

Before the reinforcement 140 . 150 and a transmission over the antenna 160 becomes a polarity code 130 applied to the pulse trains.

receiver structure

wobei α_l und τ_l jeweils ein Schwundkoeffizient und ein Delay des l.ten Pfades und n(t) ein weißer Gaußscher Prozess mit Null-Mittel und Einheitsspektraldichte ist.In a receiver 200 according to 2 is a received signal corresponding to the above-mentioned transmitted signal 201 in a single-user reusable environment:

where α _l and τ _{l are} each a fading coefficient and a delay of the lth path and n (t) is a white Gaussian process with zero mean and unit spectral density.

The receiver 200 includes a group of matched filters 210 , one for each branch. The number of branches corresponds to the number of pulse types. Each filter uses one of the corresponding template _signals s _temp (-t) 411 - 413 , Each filter is with a corresponding delay 220 and multipliers 230 connected. An integrator 250 is used to a signed 260 estimated value 209 of the output symbol.

Because different branches for Different types of impulses are used, it is now possible to have different ones To use combination schemes, such as combining or the weighting of equal transmission factors according to the types of impulses, and the difficult decisions and majority decisions for contributions from different Adapting pulse types.

wobei das O. Bit ohne Verlust der Allgemeingültigkeit betrachtet wird.The received signal 201 is first matched by the matched filters 210 guided. Each filter is tuned to a different pulse shape as determined by the corresponding template signal. The template signal for the ^nth filter is:

where the O. bit is considered without loss of generality.

The exit 211 of the filter on the nth branch y n (t) = ∫r (τ) s temp, n (τ - t) dτ (5)

werden kombiniert 240. Wenn für alle n T_n = N_pT_f ist, können wir die Summe der Entscheidungsvariablen ausdrücken als:

Then the output of the nth filter with a delayed version 212 multiplied 230 , and all of the decision variables

are combined 240 , If, for all n, T _n = N _p T _f , we can express the sum of the decision variables as:

wobei die Beiträge von unterschiedlichen Zweigen genauso gut in Abhängigkeit von einigen Faktoren, wie beispielsweise der Impulsform, gewichtet werden können.From this we can only do a single integral operation 250 use. In Equation (6), Q is an integer specifying the integration interval. Finally, rate 250 we the signed one 260 Information bit or transmitted data symbol 209 when

where the contributions from different branches may as well be weighted depending on some factors, such as the pulse shape.

verwendet wird; und eine Majoritätsregel, Mittelwertbildung oder irgendein anderer Entscheidungsprozess 320 können genutzt werden, um das Datensymbol 209 zu bewerten.Alternatively, as in 3 is shown, a decision variable is determined in each branch 310 be by

is used; and a majority rule, averaging or any other decision making process 320 can be used to the data icon 209 to rate.

When Advantage causes the system according to the invention resistance across from Multi-access interference and inter-frame interference because of different Pulse shapes with good cross-correlation properties can be used can.

Depending on the effectiveness of the noise components and the complexity constraints, an optimal N _p for the system can be selected.

Even though the invention has been described by examples of preferred embodiments is understandable Be that different other adjustments and modifications within the scope of the invention as defined by the appended claims is, can be made.

Claims (5)

Method for signaling in a communication system with ultra-wide bandwidth, which has: Generating a Data symbol; Transfer a set of reference pulses for the data symbol, each one Reference pulse of a different type; and Transfer a data pulse for each reference pulse, one type of each data pulse being identical with the type of the corresponding reference pulse. The method of claim 1, wherein each set of Reference pulses and corresponding data pulses transmitted several times becomes. The method of claim 1, wherein the pulse types selected are from the group consisting of Gaussian, Gaussian monocycle, a first Derivation of the Gaussian Monocycle, a second derivative of the Gaussian monocycle, a third Derivation of the Gaussian Monocycle, and a Gaussian Doublet. The method of claim 1, further comprising: Receive the set of reference pulses and the set of data pulses; Filter each reference pulse according to the adjustment to a template pulse; Multiply each received Data pulses with a delayed Version of the received data pulse to the data symbol again manufacture. The method of claim 4, wherein a received data pulse is weighted according to the pulse type becomes.